Onyx Security's $113M Haul Signals Rise of the AI Control Plane

NEW YORK & TEL AVIV – July 29, 2026 – In a move that sends a clear signal about the future of enterprise technology, AI control company Onyx Security has secured a massive $113 million Series B funding round. The investment, led by Bessemer Venture Partners, underscores a growing, urgent consensus in the tech world: as artificial intelligence becomes more autonomous, someone needs to be in control.

Just four months after emerging from stealth, the two-year-old firm has seen its revenue quadruple, a testament to the surging demand for its solution. Onyx is not just another cybersecurity startup; it's pioneering what it calls a "secure AI control plane," a new layer of infrastructure designed to manage the powerful, and potentially perilous, AI agents being unleashed within corporate walls. This funding brings the company’s total raised to $153 million and values it at an estimated $640 million, positioning it as a leader in a category that many believe will define the next decade of enterprise security.

The New Frontier of Risk: The Agentic Era

The excitement and anxiety surrounding AI have reached a fever pitch. Enterprises are no longer just experimenting with chatbots or predictive analytics. They are deploying sophisticated AI "agents"—autonomous actors capable of executing complex, multi-step tasks, from writing and deploying code to managing critical business workflows. This leap in capability, however, introduces what Onyx CEO Maxim Bar Kogan calls a "fundamental new risk."

"In 2025, less than one percent of actions were made by agents,” Bar Kogan noted. “These new autonomous actors will soon be taking the overwhelming majority of actions, which begs the question - how do we guarantee these actions are legitimate?”

The problem is that these agents are empowered but not accountable. They are given access to sensitive data, internal systems, and cloud infrastructure, creating a vast and unpredictable new attack surface. Traditional security tools, built for human-driven workflows and predictable applications, are ill-equipped to handle this "agentic era." The result is a critical governance gap. Research highlights the scale of the issue, indicating that a majority of enterprises are already exposing sensitive data, running agents with excessive permissions, and facing new code execution vulnerabilities through these AI systems.

Building the AI Control Plane

This is the challenge Onyx Security was built to solve. The company offers a platform that provides a unified system for discovering, monitoring, and controlling every AI agent operating within an organization. It's not about blocking AI, but enabling its safe adoption.

The platform works by continuously scanning an organization's environment—from cloud infrastructure to developer endpoints—to identify all active AI agents. From there, its proprietary models and supervisory agents monitor every step of an agent's reasoning process in real-time. If an agent attempts to perform an action that violates company policy, accesses unauthorized data, or behaves anomalously, the control plane can intervene. It can block the action, require human approval, or gently steer the agent toward a safer course of action, all based on policies that can be defined in simple natural language.

The market is already validating this approach. The firm’s rapid revenue growth and its integration with major AI providers, including a June partnership with Anthropic to secure enterprise AI adoption, demonstrate powerful market traction. For customers, the value is clear. “Onyx gives us the visibility and control we need to enable safe use of coding agents across multiple service providers,” explained Uros Solar, Head of Security Engineering at Revolut. “The platform’s unique approach allows us to evaluate and stop threats at runtime.”

The Investor Bet on AI Accountability

The $113 million investment is more than just fuel for Onyx; it's a major wager by some of the world's top venture firms on the emergence of AI control as a foundational technology category. The round saw participation from a who's who of tech investors, including Cyberstarts, TCV, Conviction, and FirstMark.

“AI agents are becoming an important part of how modern companies operate, but enterprises need trust, control, and accountability before they can adopt them at scale,” said Amit Karp, Partner at Bessemer Venture Partners. “Onyx is building the control layer that makes this possible.”

This sentiment is echoed by other key backers. Hila Zigman, General Partner at Cyberstarts, which led Onyx’s seed round, sees this as a paradigm shift. "As AI agents become embedded in critical business workflows, enterprises need a way to ensure they operate safely, predictably, and within policy," she stated. "We believe this will become one of the defining security categories of the coming decade.”

This investor confidence is bolstered by market forecasts. The emerging AI Systems Security (AISS) market is projected to skyrocket from virtually zero to nearly $8 billion by 2030. Investors see a future where the trillions being poured into AI development must be matched by a robust infrastructure for safety and governance. They are betting that the companies providing this crucial layer will become as fundamental as the cloud platforms of the previous decade.

Charting a Course for Safe AI

With its war chest replenished, Onyx plans to accelerate its mission. The new capital will be used to train the next generation of its proprietary supervisory models and to aggressively scale its go-to-market operations, particularly in the United States. The company, which already boasts a team of over 80 people across its offices in Tel Aviv and New York, is poised for significant expansion.

The founders, CEO Maxim Bar Kogan and Chief AI Officer Gil Elbaz, bring a unique blend of elite cybersecurity experience from the Israeli Defense Forces' Unit 8200 and deep AI research expertise from companies like NVIDIA. This combination of security-first thinking and advanced AI knowledge is central to the company’s ability to tackle this complex, multidisciplinary problem.

As enterprises race to integrate ever more powerful models, the question is shifting from if they will adopt AI to how they will manage the inherent risks. By providing a framework for accountability, Onyx Security is not just building a security product; it is attempting to build the trust layer necessary for the entire AI ecosystem to move forward, bridging the critical gap between ambition and secure, responsible implementation.